Biofeedback, Relaxation and Digital Training Emerge as Drug-Free Options for Children’s Headaches

Primary headaches—migraine and tension-type headache chief among them—are among the most common neurological complaints of childhood and adolescence, and a strikingly large share of affected young people never receive a formal diagnosis or targeted treatment. For families wary of putting a growing brain on daily preventive medication, the question of whether behavioral and mind–body therapies actually work has long hovered in a fog of small trials and conflicting reviews. A new systematic review and network meta-analysis published in the World Journal of Pediatrics now offers one of the most detailed comparative maps to date, ranking non-drug interventions head to head across three clinically meaningful outcomes: how often headaches strike, how long they last, and how severe they become.

The research team, led by Fa-Qiang Zhang and colleagues at Lanzhou University in collaboration with clinicians in Zhengzhou and Lanzhou, China, systematically searched PubMed, the Cochrane Library, Web of Science, and Embase from database inception through August 1, 2026. Their protocol was registered in advance in PROSPERO, and they applied the Cochrane risk-of-bias tool to each randomized controlled trial they found. In total, sixteen randomized controlled trials encompassing 820 children and adolescents met the inclusion criteria, spanning interventions as diverse as biofeedback, relaxation training, cognitive behavioral therapy, therapist-led group training, headache education, a self-guided CD-ROM program called Headstrong, music therapy, hypnotherapy, and meditation.

What makes this analysis methodologically interesting is its use of network meta-analysis, a statistical framework that goes beyond the traditional pairwise comparison of two treatments. In a standard meta-analysis, only trials that directly compare the same two interventions can be pooled. Network meta-analysis instead builds a connected web of evidence, allowing treatments that were never tested against each other in a single trial to be compared indirectly through their shared comparators—in this case, often a waiting list control group. The team used Stata 17 to generate network diagrams, league tables, and surface under the cumulative ranking curve (SUCRA) values, a metric that expresses the probability that each treatment ranks highest for a given outcome.

The headline finding concerns headache frequency, the outcome most families and clinicians care about when considering preventive therapy. Compared with waiting list controls, biofeedback produced a standardized mean difference of −1.11 (95% confidence interval −2.04 to −0.17), therapist-assisted relaxation yielded −1.06 (95% CI −2.04 to −0.07), and relaxation training alone achieved −0.69 (95% CI −1.21 to −0.17), all indicating meaningful reductions in how frequently headaches occurred. Biofeedback topped the frequency ranking with a SUCRA value of 81.4%, meaning that across the simulated ranking distribution it had the highest probability of being the most effective intervention for cutting down headache days.

Biofeedback itself deserves a technical explanation, because its prominence in these results is not accidental. The technique teaches children to gain voluntary control over physiological processes normally considered automatic—skin temperature, muscle tension in the scalp and neck, or blood flow patterns—by displaying those signals in real time on a monitor. A child watching a thermometer trace rise as they relax learns, through repeated feedback loops, to reproduce the physiological state associated with reduced headache activity. Pediatric trials of this approach stretch back decades, including early autogenic feedback studies from the 1980s and 1990s, and the new network analysis is consistent with an earlier pediatric meta-analysis published in Pediatrics in 2016 that found biofeedback to be a promising prophylactic for childhood migraine.

On headache duration, the most striking result came from an unexpected quarter: the Headstrong CD-ROM, a structured, self-administered behavioral program, achieved a standardized mean difference of −1.63 (95% CI −2.95 to −0.30) against waiting list, the largest effect size recorded anywhere in the analysis, with a SUCRA ranking of 80.5%. Combined biofeedback-plus-relaxation training followed closely at −1.54 (95% CI −2.95 to −0.13), while relaxation alone shortened headache episodes at −0.91 (95% CI −1.61 to −0.21). The strong showing of a digital, self-guided format is notable because it hints that scalable, low-cost delivery—no therapist required, no waiting room, no travel—may be able to deliver benefits comparable to hands-on care, a proposition with obvious appeal for stretched pediatric services.

For headache intensity, the field widened considerably. Education about headaches alone produced a standardized mean difference of −0.96 (95% CI −1.72 to −0.19) and claimed the top intensity ranking with a SUCRA of 75.7%. Biofeedback (−0.85), therapist-administered group training (−0.78), cognitive behavioral therapy (−0.76), and relaxation (−0.63) all showed statistically significant improvements over control as well. That simple structured education—teaching children and parents what a primary headache is, what triggers it, and how to respond—could rank so highly is one of the more provocative implications of the study, though the authors are careful to frame these rankings as relative probabilities rather than proof of definitive clinical superiority.

The authors are equally candid about the limitations, and their caution is warranted. The certainty of evidence, assessed with the Confidence in Network Meta-Analysis (CINeMA) framework, was rated as very low across most comparisons, and many effect estimates were imprecise, with confidence intervals wide enough to accommodate both clinically trivial and substantially beneficial effects. Sixteen trials and 820 participants is a modest evidence base from which to rank more than a dozen interventions, and the underlying trials varied in age—some dating to the 1980s—in diagnostic criteria, outcome measurement, and comparator choice. Standardized mean differences, while necessary when studies use different scales, are harder to translate into the language parents actually want: how many fewer headache days per month, or how many minutes shorter an episode will be.

Context matters for interpreting why this work matters now. Epidemiological reviews have established that primary headaches affect a substantial proportion of school-age children and adolescents worldwide, and studies cited by the authors link pediatric headache to reduced school performance, degraded quality of life, and elevated rates of anxiety and depressive symptoms. Yet pharmacological preventives approved and tested specifically for children remain limited, and behavioral therapies have historically been recommended on the basis of evidence too fragmented to say which one a clinician should try first. Previous network meta-analyses in adults and one earlier pediatric analysis in Pediatrics had begun to organize this landscape, but the new World Journal of Pediatrics review extends it with more recent trials and a formal certainty assessment.

The practical takeaway is a nuanced one. For a child whose primary problem is frequent headaches, the analysis points toward biofeedback and therapist-assisted relaxation as the best-supported behavioral options. For a child whose episodes are long and disruptive, self-guided digital programs such as Headstrong or combined biofeedback-relaxation protocols deserve consideration, with the caveat that the evidence base is thin and the CD-ROM format itself feels like a relic of an earlier digital era—its modern successors, internet-delivered cognitive behavioral programs, are represented in the trial pool but did not dominate the rankings. For children whose main complaint is pain intensity, even structured education may deliver measurable relief. None of these findings should be read as a green light to abandon medical evaluation; the authors themselves describe their conclusions as exploratory. What the analysis does establish is that the question of drug-free headache care in children is now answerable in comparative, quantitative terms—and that the answer, pending better trials, leans encouragingly toward the behavioral side of the clinic.

Subject of Research: Comparative efficacy of non-pharmacological interventions for pediatric primary headaches

Article Title: Efficacy of non-pharmacological interventions for pediatric primary headaches: a systematic review and network meta-analysis

Article References: Zhang, F.-Q., Wu, Y.-N., Liu, W.-D., Yang, K.-H., Li, X.-X., & Wang, S.-Y. (2026). Efficacy of non-pharmacological interventions for pediatric primary headaches: a systematic review and network meta-analysis. World Journal of Pediatrics. https://doi.org/10.1007/s12519-026-01103-9

Image Credits: AI Generated

DOI: 10.1007/s12519-026-01103-9

Keywords: pediatric headache, migraine, tension-type headache, biofeedback, relaxation training, cognitive behavioral therapy, network meta-analysis, non-pharmacological interventions, children, adolescents, Headstrong CD-ROM, evidence certainty